crewAI/lib/crewai/tests/llms/openai/test_openai.py

import json
import os
import sys
import types
from typing import Any
from unittest.mock import patch, MagicMock
import openai
import pytest

from crewai.llm import CONTEXT_WINDOW_USAGE_RATIO, LLM
from crewai.llms.providers.openai.completion import OpenAICompletion, ResponsesAPIResult
from crewai.crew import Crew
from crewai.agent import Agent
from crewai.task import Task

def test_openai_completion_is_used_when_openai_provider():
    """
    Test that OpenAICompletion from completion.py is used when LLM uses provider 'openai'
    """
    llm = LLM(model="gpt-4o")

    assert llm.__class__.__name__ == "OpenAICompletion"
    assert llm.provider == "openai"
    assert llm.model == "gpt-4o"


def test_openai_completion_is_used_when_no_provider_prefix():
    """
    Test that OpenAICompletion is used when no provider prefix is given (defaults to openai)
    """
    llm = LLM(model="gpt-4o")

    from crewai.llms.providers.openai.completion import OpenAICompletion
    assert llm.__class__.__name__ == "OpenAICompletion"
    assert llm.provider == "openai"
    assert llm.model == "gpt-4o"


def test_custom_openai_flag_uses_native_openai_without_provider_prefix():
    """Custom OpenAI-compatible endpoints can serve arbitrary model ids."""
    with patch.dict(os.environ, {"OPENAI_API_KEY": "test-key"}, clear=False):
        llm = LLM(
            model="anthropic/claude-sonnet-4-6",
            custom_openai=True,
            base_url="https://gateway.example/v1",
            is_litellm=False,
        )

    assert llm.__class__.__name__ == "OpenAICompletion"
    assert llm.is_litellm is False
    assert llm.provider == "openai"
    assert llm.model == "anthropic/claude-sonnet-4-6"
    assert llm.base_url == "https://gateway.example/v1"
    assert llm.custom_openai is True
    assert "custom_openai" not in llm.additional_params

    config = llm.to_config_dict()
    assert config["model"] == "anthropic/claude-sonnet-4-6"
    assert config["custom_openai"] is True
    assert config["base_url"] == "https://gateway.example/v1"

    rebuilt = LLM(**config)
    assert isinstance(rebuilt, OpenAICompletion)
    assert rebuilt.model == "anthropic/claude-sonnet-4-6"
    assert rebuilt.base_url == "https://gateway.example/v1"


def test_custom_openai_flag_requires_custom_base_url():
    """Avoid routing arbitrary custom model ids to api.openai.com by mistake."""
    with patch.dict(os.environ, {"OPENAI_API_KEY": "test-key"}, clear=True):
        with pytest.raises(ValueError, match="custom_openai=True requires"):
            LLM(
                model="anthropic/claude-sonnet-4-6",
                custom_openai=True,
                is_litellm=False,
            )


def test_direct_custom_openai_completion_requires_custom_base_url():
    """Direct construction must not silently fall back to api.openai.com."""
    with patch.dict(os.environ, {"OPENAI_API_KEY": "test-key"}, clear=True):
        with pytest.raises(ValueError, match="custom_openai=True requires"):
            OpenAICompletion(
                model="anthropic/claude-sonnet-4-6",
                custom_openai=True,
            )


def test_custom_openai_flag_strips_openai_routing_prefix():
    """The openai/ routing prefix is not part of the gateway's model id."""
    with patch.dict(os.environ, {"OPENAI_API_KEY": "test-key"}, clear=False):
        llm = LLM(
            model="openai/anthropic/claude-sonnet-4-6",
            custom_openai=True,
            base_url="https://gateway.example/v1",
            is_litellm=False,
        )

    assert isinstance(llm, OpenAICompletion)
    assert llm.model == "anthropic/claude-sonnet-4-6"


def test_openai_prefixed_custom_endpoint_uses_native_sdk_for_nested_model_id():
    """Custom OpenAI-compatible endpoints may serve non-OpenAI model ids."""
    with patch.dict(os.environ, {"OPENAI_API_KEY": "test-key"}, clear=False):
        llm = LLM(
            model="openai/anthropic/claude-sonnet-4-6",
            base_url="https://gateway.example/v1",
            is_litellm=False,
        )

    assert llm.__class__.__name__ == "OpenAICompletion"
    assert llm.is_litellm is False
    assert llm.provider == "openai"
    assert llm.model == "anthropic/claude-sonnet-4-6"
    assert llm.custom_openai is True
    assert llm.base_url == "https://gateway.example/v1"

def test_explicit_custom_openai_uses_legacy_api_base_env_var():
    """Explicit custom routing supports the legacy endpoint environment variable."""
    with patch.dict(
        os.environ,
        {
            "OPENAI_API_KEY": "test-key",
            "OPENAI_API_BASE": "https://gateway.example/v1",
        },
        clear=False,
    ):
        os.environ.pop("OPENAI_BASE_URL", None)
        llm = LLM(
            model="openai/anthropic/claude-sonnet-4-6",
            custom_openai=True,
            is_litellm=False,
        )

    assert isinstance(llm, OpenAICompletion)
    assert llm.is_litellm is False
    assert llm.provider == "openai"
    assert llm.model == "anthropic/claude-sonnet-4-6"
    assert llm.custom_openai is True


def test_openai_prefixed_unknown_model_ignores_ambient_base_url_for_routing():
    """Ambient OpenAI configuration must not opt unknown models into native routing."""
    with patch.dict(
        os.environ,
        {
            "OPENAI_API_KEY": "test-key",
            "OPENAI_BASE_URL": "https://gateway.example/v1",
        },
        clear=True,
    ):
        with (
            patch("crewai.llm._ensure_litellm", return_value=False),
            pytest.raises(ImportError, match="LiteLLM fallback package"),
        ):
            LLM(model="openai/not-a-real-openai-model")


@pytest.mark.parametrize("endpoint_field", ["api_base", "env"])
def test_custom_openai_config_preserves_resolved_endpoint(endpoint_field):
    """Serialized custom OpenAI configs can reconstruct the same endpoint."""
    kwargs = {}
    env = {"OPENAI_API_KEY": "test-key"}
    if endpoint_field == "api_base":
        kwargs["api_base"] = "https://gateway.example/v1"
    else:
        env["OPENAI_API_BASE"] = "https://gateway.example/v1"

    with patch.dict(os.environ, env, clear=True):
        llm = LLM(
            model="anthropic/claude-sonnet-4-6",
            custom_openai=True,
            **kwargs,
        )

    config = llm.to_config_dict()
    assert config["base_url"] == "https://gateway.example/v1"
    rebuilt = LLM(**config)
    assert isinstance(rebuilt, OpenAICompletion)
    assert rebuilt.base_url == "https://gateway.example/v1"


@pytest.mark.vcr()
def test_openai_is_default_provider_without_explicit_llm_set_on_agent():
    """
    Test that OpenAI is the default provider when no explicit LLM is set on the agent
    """
    agent = Agent(
        role="Research Assistant",
        goal="Find information about the population of Tokyo",
        backstory="You are a helpful research assistant.",
        llm=LLM(model="gpt-4o-mini"),
    )
    task = Task(
        description="Find information about the population of Tokyo",
        expected_output="The population of Tokyo is 10 million",
        agent=agent,
    )
    crew = Crew(agents=[agent], tasks=[task])
    crew.kickoff()
    assert crew.agents[0].llm.__class__.__name__ == "OpenAICompletion"
    assert crew.agents[0].llm.model == "gpt-4o-mini"






def test_openai_completion_module_is_imported(monkeypatch):
    """
    Test that the completion module is properly imported when using OpenAI provider
    """
    module_name = "crewai.llms.providers.openai.completion"

    monkeypatch.delitem(sys.modules, module_name, raising=False)

    LLM(model="gpt-4o")

    assert module_name in sys.modules
    completion_mod = sys.modules[module_name]
    assert isinstance(completion_mod, types.ModuleType)

    assert hasattr(completion_mod, 'OpenAICompletion')


def test_native_openai_raises_error_when_initialization_fails():
    """
    Test that LLM raises ImportError when native OpenAI completion fails to initialize.
    This ensures we don't silently fall back when there's a configuration issue.
    """
    with patch('crewai.llm.LLM._get_native_provider') as mock_get_provider:

        class FailingCompletion:
            def __init__(self, *args, **kwargs):
                raise Exception("Native SDK failed")

        mock_get_provider.return_value = FailingCompletion

        # This should raise ImportError, not fall back to LiteLLM
        with pytest.raises(ImportError) as excinfo:
            LLM(model="gpt-4o")

        assert "Error importing native provider" in str(excinfo.value)
        assert "Native SDK failed" in str(excinfo.value)


def test_openai_completion_initialization_parameters():
    """
    Test that OpenAICompletion is initialized with correct parameters
    """
    llm = LLM(
        model="gpt-4o",
        temperature=0.7,
        max_tokens=1000,
        api_key="test-key"
    )

    from crewai.llms.providers.openai.completion import OpenAICompletion
    assert isinstance(llm, OpenAICompletion)
    assert llm.model == "gpt-4o"
    assert llm.temperature == 0.7
    assert llm.max_tokens == 1000

def test_openai_completion_call():
    """
    Test that OpenAICompletion call method works
    """
    llm = LLM(model="openai/gpt-4o")

    with patch.object(llm, 'call', return_value="Hello! I'm ready to help.") as mock_call:
        result = llm.call("Hello, how are you?")

        assert result == "Hello! I'm ready to help."
        mock_call.assert_called_once_with("Hello, how are you?")


def test_openai_completion_called_during_crew_execution():
    """
    Test that OpenAICompletion.call is actually invoked when running a crew
    """
    openai_llm = LLM(model="openai/gpt-4o")

    with patch.object(openai_llm, 'call', return_value="Tokyo has 14 million people.") as mock_call:

        agent = Agent(
            role="Research Assistant",
            goal="Find population info",
            backstory="You research populations.",
            llm=openai_llm,
        )

        task = Task(
            description="Find Tokyo population",
            expected_output="Population number",
            agent=agent,
        )

        crew = Crew(agents=[agent], tasks=[task])
        result = crew.kickoff()

        assert mock_call.called
        assert "14 million" in str(result)


def test_openai_completion_call_arguments():
    """
    Test that OpenAICompletion.call is invoked with correct arguments
    """
    openai_llm = LLM(model="openai/gpt-4o")

    with patch.object(openai_llm, 'call') as mock_call:
        mock_call.return_value = "Task completed successfully."

        agent = Agent(
            role="Test Agent",
            goal="Complete a simple task",
            backstory="You are a test agent.",
            llm=openai_llm
        )

        task = Task(
            description="Say hello world",
            expected_output="Hello world",
            agent=agent,
        )

        crew = Crew(agents=[agent], tasks=[task])
        crew.kickoff()

        assert mock_call.called

        call_args = mock_call.call_args
        assert call_args is not None

        messages = call_args[0][0]
        assert isinstance(messages, (str, list))

        if isinstance(messages, str):
            assert "hello world" in messages.lower()
        elif isinstance(messages, list):
            message_content = str(messages).lower()
            assert "hello world" in message_content


def test_multiple_openai_calls_in_crew():
    """
    Test that OpenAICompletion.call is invoked multiple times for multiple tasks
    """
    openai_llm = LLM(model="openai/gpt-4o")

    with patch.object(openai_llm, 'call') as mock_call:
        mock_call.return_value = "Task completed."

        agent = Agent(
            role="Multi-task Agent",
            goal="Complete multiple tasks",
            backstory="You can handle multiple tasks.",
            llm=openai_llm
        )

        task1 = Task(
            description="First task",
            expected_output="First result",
            agent=agent,
        )

        task2 = Task(
            description="Second task",
            expected_output="Second result",
            agent=agent,
        )

        crew = Crew(
            agents=[agent],
            tasks=[task1, task2]
        )
        crew.kickoff()

        assert mock_call.call_count >= 2  # At least one call per task

        for call in mock_call.call_args_list:
            assert len(call[0]) > 0
            messages = call[0][0]
            assert messages is not None


def test_openai_completion_with_tools():
    """
    Test that OpenAICompletion.call is invoked with tools when agent has tools
    """
    from crewai.tools import tool

    @tool
    def sample_tool(query: str) -> str:
        """A sample tool for testing"""
        return f"Tool result for: {query}"

    openai_llm = LLM(model="openai/gpt-4o")

    with patch.object(openai_llm, 'call') as mock_call:
        mock_call.return_value = "Task completed with tools."

        agent = Agent(
            role="Tool User",
            goal="Use tools to complete tasks",
            backstory="You can use tools.",
            llm=openai_llm,
            tools=[sample_tool]
        )

        task = Task(
            description="Use the sample tool",
            expected_output="Tool usage result",
            agent=agent,
        )

        crew = Crew(agents=[agent], tasks=[task])
        crew.kickoff()

        assert mock_call.called

        call_args = mock_call.call_args
        call_kwargs = call_args[1] if len(call_args) > 1 else {}

        if 'tools' in call_kwargs:
            assert call_kwargs['tools'] is not None
            assert len(call_kwargs['tools']) > 0

@pytest.mark.vcr()
def test_openai_completion_call_returns_usage_metrics():
    """
    Test that OpenAICompletion.call returns usage metrics
    """
    agent = Agent(
        role="Research Assistant",
        goal="Find information about the population of Tokyo",
        backstory="You are a helpful research assistant.",
        llm=LLM(model="gpt-4o"),
        verbose=True,
    )

    task = Task(
        description="Find information about the population of Tokyo",
        expected_output="The population of Tokyo is 10 million",
        agent=agent,
    )

    crew = Crew(agents=[agent], tasks=[task])
    result = crew.kickoff()
    assert result.token_usage is not None
    assert result.token_usage.total_tokens == 289
    assert result.token_usage.prompt_tokens == 173
    assert result.token_usage.completion_tokens == 116
    assert result.token_usage.successful_requests == 1
    assert result.token_usage.cached_prompt_tokens == 0


@pytest.mark.skip(reason="Allow for litellm")
def test_openai_raises_error_when_model_not_supported():
    """Test that OpenAICompletion raises ValueError when model not supported"""

    with patch('crewai.llms.providers.openai.completion.OpenAI') as mock_openai_class:
        mock_client = MagicMock()
        mock_openai_class.return_value = mock_client

        mock_client.chat.completions.create.side_effect = openai.NotFoundError(
            message="The model `model-doesnt-exist` does not exist",
            response=MagicMock(),
            body={}
        )

        llm = LLM(model="openai/model-doesnt-exist")

        with pytest.raises(ValueError, match="Model.*not found"):
            llm.call("Hello")


def _raw_create_double(content: str = "test response") -> MagicMock:
    """Double for `chat.completions.with_raw_response.create`.

    `text` has to be real JSON carrying `choices` so the gateway error-envelope
    guard sees a well-formed completion and defers to `parse()`.
    """
    parsed = MagicMock(
        choices=[MagicMock(message=MagicMock(content=content, tool_calls=None))],
        usage=MagicMock(prompt_tokens=10, completion_tokens=20, total_tokens=30),
    )
    return MagicMock(
        text=json.dumps({"choices": [{"index": 0}]}),
        **{"parse.return_value": parsed},
    )


def test_openai_client_setup_with_extra_arguments():
    """
    Test that OpenAICompletion is initialized with correct parameters
    """
    llm = LLM(
        model="gpt-4o",
        temperature=0.7,
        max_tokens=1000,
        top_p=0.5,
        max_retries=3,
        timeout=30
    )

    assert llm.temperature == 0.7
    assert llm.max_tokens == 1000
    assert llm.top_p == 0.5

    assert llm._client.max_retries == 3
    assert llm._client.timeout == 30

    with patch.object(
        llm._client.chat.completions.with_raw_response, 'create'
    ) as mock_create:
        mock_create.return_value = _raw_create_double()

        llm.call("Hello")

        call_args = mock_create.call_args[1]
        assert call_args['temperature'] == 0.7
        assert call_args['max_tokens'] == 1000
        assert call_args['top_p'] == 0.5
        assert call_args['model'] == 'gpt-4o'

def test_extra_arguments_are_passed_to_openai_completion():
    """
    Test that extra arguments are passed to OpenAICompletion
    """
    llm = LLM(model="gpt-4o", temperature=0.7, max_tokens=1000, top_p=0.5, max_retries=3)

    with patch.object(
        llm._client.chat.completions.with_raw_response, 'create'
    ) as mock_create:
        mock_create.return_value = _raw_create_double()

        llm.call("Hello, how are you?")

        assert mock_create.called
        call_kwargs = mock_create.call_args[1]

        assert call_kwargs['temperature'] == 0.7
        assert call_kwargs['max_tokens'] == 1000
        assert call_kwargs['top_p'] == 0.5
        assert call_kwargs['model'] == 'gpt-4o'



def test_openai_get_client_params_with_api_base():
    """
    Test that _get_client_params correctly converts api_base to base_url
    """
    llm = OpenAICompletion(
        model="gpt-4o",
        api_base="https://custom.openai.com/v1",
    )
    client_params = llm._get_client_params()
    assert client_params["base_url"] == "https://custom.openai.com/v1"

def test_openai_get_client_params_with_base_url_priority():
    """
    Test that base_url takes priority over api_base in _get_client_params
    """
    llm = OpenAICompletion(
        model="gpt-4o",
        base_url="https://priority.openai.com/v1",
        api_base="https://fallback.openai.com/v1",
    )
    client_params = llm._get_client_params()
    assert client_params["base_url"] == "https://priority.openai.com/v1"

def test_openai_get_client_params_with_env_var():
    """
    Test that _get_client_params uses OPENAI_BASE_URL environment variable as fallback
    """
    with patch.dict(os.environ, {
        "OPENAI_BASE_URL": "https://env.openai.com/v1",
    }):
        llm = OpenAICompletion(model="gpt-4o")
        client_params = llm._get_client_params()
        assert client_params["base_url"] == "https://env.openai.com/v1"

def test_openai_get_client_params_with_legacy_api_base_env_var():
    """
    Test that _get_client_params uses OPENAI_API_BASE when OPENAI_BASE_URL is absent.
    """
    with patch.dict(os.environ, {
        "OPENAI_API_BASE": "https://legacy-env.openai.com/v1",
    }, clear=False):
        os.environ.pop("OPENAI_BASE_URL", None)
        llm = OpenAICompletion(model="gpt-4o")
        client_params = llm._get_client_params()
        assert client_params["base_url"] == "https://legacy-env.openai.com/v1"

def test_openai_get_client_params_priority_order():
    """
    Test the priority order: base_url > api_base > OPENAI_BASE_URL > OPENAI_API_BASE
    """
    with patch.dict(os.environ, {
        "OPENAI_BASE_URL": "https://env.openai.com/v1",
        "OPENAI_API_BASE": "https://legacy-env.openai.com/v1",
    }):
        llm1 = OpenAICompletion(
            model="gpt-4o",
            base_url="https://base-url.openai.com/v1",
            api_base="https://api-base.openai.com/v1",
        )
        params1 = llm1._get_client_params()
        assert params1["base_url"] == "https://base-url.openai.com/v1"

        llm2 = OpenAICompletion(
            model="gpt-4o",
            api_base="https://api-base.openai.com/v1",
        )
        params2 = llm2._get_client_params()
        assert params2["base_url"] == "https://api-base.openai.com/v1"

        llm3 = OpenAICompletion(model="gpt-4o")
        params3 = llm3._get_client_params()
        assert params3["base_url"] == "https://env.openai.com/v1"

def test_openai_get_client_params_no_base_url(monkeypatch):
    """
    Test that _get_client_params works correctly when no base_url is specified
    """
    monkeypatch.delenv("OPENAI_BASE_URL", raising=False)
    monkeypatch.delenv("OPENAI_API_BASE", raising=False)

    llm = OpenAICompletion(model="gpt-4o")
    client_params = llm._get_client_params()
    # When no base_url is provided, it should not be in the params (filtered out as None)
    assert "base_url" not in client_params or client_params.get("base_url") is None


def test_openai_streaming_with_response_model():
    """
    Test that streaming with response_model works correctly and doesn't call invalid API methods.
    This test verifies the fix for the bug where streaming with response_model attempted to call
    self.client.responses.stream() with invalid parameters (input, text_format).
    """
    from pydantic import BaseModel

    class TestResponse(BaseModel):
        """Test response model."""

        answer: str
        confidence: float

    llm = LLM(model="openai/gpt-4o", stream=True)

    with patch.object(llm._client.beta.chat.completions, "stream") as mock_stream:
        mock_chunk1 = MagicMock()
        mock_chunk1.type = "content.delta"
        mock_chunk1.delta = '{"answer": "test", '
        mock_chunk1.id = "response-1"

        mock_chunk2 = MagicMock()
        mock_chunk2.type = "content.delta"
        mock_chunk2.delta = '"confidence": 0.95}'
        mock_chunk2.id = "response-2"

        mock_parsed = TestResponse(answer="test", confidence=0.95)
        mock_message = MagicMock()
        mock_message.parsed = mock_parsed
        mock_choice = MagicMock()
        mock_choice.message = mock_message
        mock_final_completion = MagicMock()
        mock_final_completion.choices = [mock_choice]

        mock_stream_obj = MagicMock()
        mock_stream_obj.__enter__ = MagicMock(return_value=mock_stream_obj)
        mock_stream_obj.__exit__ = MagicMock(return_value=None)
        mock_stream_obj.__iter__ = MagicMock(return_value=iter([mock_chunk1, mock_chunk2]))
        mock_stream_obj.get_final_completion = MagicMock(return_value=mock_final_completion)

        mock_stream.return_value = mock_stream_obj

        result = llm.call("Test question", response_model=TestResponse)

        assert result is not None
        assert isinstance(result, TestResponse)
        assert result.answer == "test"
        assert result.confidence == 0.95

        assert mock_stream.called
        call_kwargs = mock_stream.call_args[1]
        assert call_kwargs["model"] == "gpt-4o"
        assert call_kwargs["response_format"] == TestResponse

        assert "input" not in call_kwargs
        assert "text_format" not in call_kwargs


@pytest.mark.vcr()
def test_openai_response_format_with_pydantic_model():
    """
    Test that response_format with a Pydantic BaseModel returns structured output.
    """
    from pydantic import BaseModel, Field

    class AnswerResponse(BaseModel):
        """Response model with structured fields."""

        answer: str = Field(description="The answer to the question")
        confidence: float = Field(description="Confidence score between 0 and 1")

    llm = LLM(model="gpt-4o", response_format=AnswerResponse)
    result = llm.call("What is the capital of France? Be concise.")

    assert isinstance(result, AnswerResponse)
    assert result.answer is not None
    assert 0 <= result.confidence <= 1


@pytest.mark.vcr()
def test_openai_response_format_with_dict():
    """
    Test that response_format with a dict returns JSON output.
    """
    import json

    llm = LLM(model="gpt-4o", response_format={"type": "json_object"})
    result = llm.call("Return a JSON object with a 'status' field set to 'success'")

    parsed = json.loads(result)
    assert "status" in parsed


@pytest.mark.vcr()
def test_openai_response_format_none():
    """
    Test that when response_format is None, the API returns plain text.
    """
    llm = LLM(model="gpt-4o", response_format=None)
    result = llm.call("Say hello in one word")

    assert isinstance(result, str)
    assert len(result) > 0


@pytest.mark.vcr()
def test_openai_streaming_returns_usage_metrics():
    """
    Test that OpenAI streaming calls return proper token usage metrics.
    """
    agent = Agent(
        role="Research Assistant",
        goal="Find information about the capital of France",
        backstory="You are a helpful research assistant.",
        llm=LLM(model="gpt-4o-mini", stream=True),
        verbose=True,
    )

    task = Task(
        description="What is the capital of France?",
        expected_output="The capital of France",
        agent=agent,
    )

    crew = Crew(agents=[agent], tasks=[task])
    result = crew.kickoff()

    assert result.token_usage is not None
    assert result.token_usage.total_tokens > 0
    assert result.token_usage.prompt_tokens > 0
    assert result.token_usage.completion_tokens > 0
    assert result.token_usage.successful_requests >= 1


def test_openai_responses_api_initialization():
    """Test that OpenAI Responses API can be initialized with api='responses'."""
    llm = OpenAICompletion(
        model="gpt-5",
        api="responses",
        instructions="You are a helpful assistant.",
        store=True,
    )

    assert llm.api == "responses"
    assert llm.instructions == "You are a helpful assistant."
    assert llm.store is True
    assert llm.model == "gpt-5"


def test_openai_responses_api_default_is_completions():
    """Test that the default API is 'completions' for backward compatibility."""
    llm = OpenAICompletion(model="gpt-4o")

    assert llm.api == "completions"


def test_openai_responses_api_prepare_params():
    """Test that Responses API params are prepared correctly."""
    llm = OpenAICompletion(
        model="gpt-5",
        api="responses",
        instructions="Base instructions.",
        store=True,
        temperature=0.7,
    )

    messages = [
        {"role": "system", "content": "System message."},
        {"role": "user", "content": "Hello!"},
    ]

    params = llm._prepare_responses_params(messages)

    assert params["model"] == "gpt-5"
    assert "Base instructions." in params["instructions"]
    assert "System message." in params["instructions"]
    assert params["store"] is True
    assert params["temperature"] == 0.7
    assert params["input"] == [{"role": "user", "content": "Hello!"}]


def test_openai_responses_api_tool_format():
    """Test that tools are converted to Responses API format (internally-tagged)."""
    llm = OpenAICompletion(model="gpt-5", api="responses")

    tools = [
        {
            "name": "get_weather",
            "description": "Get the weather for a location",
            "parameters": {
                "type": "object",
                "properties": {"location": {"type": "string"}},
                "required": ["location"],
            },
        }
    ]

    responses_tools = llm._convert_tools_for_responses(tools)

    assert len(responses_tools) == 1
    tool = responses_tools[0]
    assert tool["type"] == "function"
    assert tool["name"] == "get_weather"
    assert tool["description"] == "Get the weather for a location"
    assert "parameters" in tool
    assert "function" not in tool


def test_openai_completions_api_tool_format():
    """Test that tools are converted to Chat Completions API format (externally-tagged)."""
    llm = OpenAICompletion(model="gpt-4o", api="completions")

    tools = [
        {
            "name": "get_weather",
            "description": "Get the weather for a location",
            "parameters": {
                "type": "object",
                "properties": {"location": {"type": "string"}},
                "required": ["location"],
            },
        }
    ]

    completions_tools = llm._convert_tools_for_interference(tools)

    assert len(completions_tools) == 1
    tool = completions_tools[0]
    assert tool["type"] == "function"
    assert "function" in tool
    assert tool["function"]["name"] == "get_weather"
    assert tool["function"]["description"] == "Get the weather for a location"


def test_openai_responses_api_structured_output_format():
    """Test that structured outputs use text.format for Responses API."""
    from pydantic import BaseModel

    class Person(BaseModel):
        name: str
        age: int

    llm = OpenAICompletion(model="gpt-5", api="responses")

    messages = [{"role": "user", "content": "Extract: Jane, 25"}]
    params = llm._prepare_responses_params(messages, response_model=Person)

    assert "text" in params
    assert "format" in params["text"]
    assert params["text"]["format"]["type"] == "json_schema"
    assert params["text"]["format"]["name"] == "Person"
    assert params["text"]["format"]["strict"] is True


def test_openai_responses_api_with_previous_response_id():
    """Test that previous_response_id is passed for multi-turn conversations."""
    llm = OpenAICompletion(
        model="gpt-5",
        api="responses",
        previous_response_id="resp_abc123",
        store=True,
    )

    messages = [{"role": "user", "content": "Continue our conversation."}]
    params = llm._prepare_responses_params(messages)

    assert params["previous_response_id"] == "resp_abc123"
    assert params["store"] is True


def test_openai_responses_api_call_routing():
    """Test that call() routes to the correct API based on the api parameter."""
    from unittest.mock import patch, MagicMock

    llm_completions = OpenAICompletion(model="gpt-4o", api="completions")
    llm_responses = OpenAICompletion(model="gpt-5", api="responses")

    with patch.object(
        llm_completions, "_call_completions", return_value="completions result"
    ) as mock_completions:
        result = llm_completions.call("Hello")
        mock_completions.assert_called_once()
        assert result == "completions result"

    with patch.object(
        llm_responses, "_call_responses", return_value="responses result"
    ) as mock_responses:
        result = llm_responses.call("Hello")
        mock_responses.assert_called_once()
        assert result == "responses result"


# VCR Integration Tests for Responses API


@pytest.mark.vcr()
def test_openai_responses_api_basic_call():
    """Test basic Responses API call with text generation."""
    llm = OpenAICompletion(
        model="gpt-4o-mini",
        api="responses",
        instructions="You are a helpful assistant. Be concise.",
    )

    result = llm.call("What is 2 + 2? Answer with just the number.")

    assert isinstance(result, str)
    assert "4" in result


@pytest.mark.vcr()
def test_openai_responses_api_with_structured_output():
    """Test Responses API with structured output using Pydantic model."""
    from pydantic import BaseModel, Field

    class MathAnswer(BaseModel):
        """Structured math answer."""

        result: int = Field(description="The numerical result")
        explanation: str = Field(description="Brief explanation")

    llm = OpenAICompletion(
        model="gpt-4o-mini",
        api="responses",
    )

    result = llm.call("What is 5 * 7?", response_model=MathAnswer)

    assert isinstance(result, MathAnswer)
    assert result.result == 35


@pytest.mark.vcr()
def test_openai_responses_api_with_system_message_extraction():
    """Test that system messages are properly extracted to instructions."""
    llm = OpenAICompletion(
        model="gpt-4o-mini",
        api="responses",
    )

    messages = [
        {"role": "system", "content": "You always respond in uppercase letters only."},
        {"role": "user", "content": "Say hello"},
    ]

    result = llm.call(messages)

    assert isinstance(result, str)
    assert result.isupper() or "HELLO" in result.upper()


def test_openai_responses_api_converts_assistant_tool_calls_message():
    """Regression: assistant messages carrying tool_calls (Chat-Completions
    shape) must become standalone function_call input items, since the
    Responses API has no message shape for an assistant tool-call turn.
    """
    llm = OpenAICompletion(model="gpt-4o-mini", api="responses")

    messages = [
        {"role": "user", "content": "Fetch https://example.com"},
        {
            "role": "assistant",
            "content": None,
            "tool_calls": [
                {
                    "id": "call_abc123",
                    "type": "function",
                    "function": {
                        "name": "fetch_page",
                        "arguments": '{"url": "https://example.com"}',
                    },
                }
            ],
        },
    ]

    params = llm._prepare_responses_params(messages)

    assert params["input"][0] == {"role": "user", "content": "Fetch https://example.com"}
    assert params["input"][1] == {
        "type": "function_call",
        "call_id": "call_abc123",
        "name": "fetch_page",
        "arguments": '{"url": "https://example.com"}',
    }


def test_openai_responses_api_preserves_assistant_content_with_tool_calls():
    """Assistant text must be retained when it accompanies tool calls."""
    llm = OpenAICompletion(model="gpt-4o-mini", api="responses")

    messages = [
        {
            "role": "assistant",
            "content": "I'll fetch that page now.",
            "tool_calls": [
                {
                    "type": "function",
                    "id": "call_fetch_page",
                    "function": {
                        "name": "fetch_page",
                        "arguments": {"url": "https://example.com"},
                    },
                }
            ],
        }
    ]

    params = llm._prepare_responses_params(messages)

    assert params["input"][0] == {
        "role": "assistant",
        "content": "I'll fetch that page now.",
    }
    assert params["input"][1]["type"] == "function_call"
    assert params["input"][1]["call_id"] == "call_fetch_page"
    assert params["input"][1]["arguments"] == '{"url": "https://example.com"}'


def test_openai_responses_api_defaults_missing_tool_call_arguments():
    """Missing or empty tool-call arguments must become a valid JSON object."""
    llm = OpenAICompletion(model="gpt-4o-mini", api="responses")

    messages = [
        {
            "role": "assistant",
            "content": None,
            "tool_calls": [
                {
                    "id": "call_no_args",
                    "type": "function",
                    "function": {"name": "ping"},
                },
                {
                    "id": "call_empty_args",
                    "type": "function",
                    "function": {"name": "ping", "arguments": ""},
                },
            ],
        }
    ]

    params = llm._prepare_responses_params(messages)

    assert params["input"][0]["arguments"] == "{}"
    assert params["input"][1]["arguments"] == "{}"


def test_openai_responses_api_converts_tool_result_message():
    """Regression: tool-role messages (Chat-Completions shape) must become
    function_call_output input items for the Responses API.
    """
    llm = OpenAICompletion(model="gpt-4o-mini", api="responses")

    messages = [
        {
            "role": "tool",
            "tool_call_id": "call_abc123",
            "name": "fetch_page",
            "content": "<html>page text</html>",
        },
    ]

    params = llm._prepare_responses_params(messages)

    assert params["input"] == [
        {
            "type": "function_call_output",
            "call_id": "call_abc123",
            "output": "<html>page text</html>",
        }
    ]


def test_openai_responses_api_multi_turn_tool_conversation_shape():
    """Regression: a full multi-turn tool-calling conversation (user ->
    assistant tool_calls -> tool result) must convert entirely into valid
    Responses API input items, with no leftover Chat-Completions-only keys
    ("tool_calls", "tool_call_id") that the Responses API would reject.
    """
    llm = OpenAICompletion(model="gpt-4o-mini", api="responses")

    messages = [
        {"role": "user", "content": "Fetch https://example.com"},
        {
            "role": "assistant",
            "content": None,
            "tool_calls": [
                {
                    "id": "call_abc123",
                    "type": "function",
                    "function": {
                        "name": "fetch_page",
                        "arguments": '{"url": "https://example.com"}',
                    },
                }
            ],
        },
        {
            "role": "tool",
            "tool_call_id": "call_abc123",
            "name": "fetch_page",
            "content": "<html>page text</html>",
        },
    ]

    params = llm._prepare_responses_params(messages)

    for item in params["input"]:
        assert "tool_calls" not in item
        assert "tool_call_id" not in item
    assert params["input"][1]["type"] == "function_call"
    assert params["input"][2]["type"] == "function_call_output"


@pytest.mark.vcr()
def test_openai_responses_api_streaming():
    """Test Responses API with streaming enabled."""
    llm = OpenAICompletion(
        model="gpt-4o-mini",
        api="responses",
        stream=True,
        instructions="Be very concise.",
    )

    result = llm.call("Count from 1 to 3, separated by commas.")

    assert isinstance(result, str)
    assert "1" in result
    assert "2" in result
    assert "3" in result


@pytest.mark.vcr()
def test_openai_responses_api_returns_usage_metrics():
    """Test that Responses API calls return proper token usage metrics."""
    llm = OpenAICompletion(
        model="gpt-4o-mini",
        api="responses",
    )

    llm.call("Say hello")

    usage = llm.get_token_usage_summary()
    assert usage.total_tokens > 0
    assert usage.prompt_tokens > 0
    assert usage.completion_tokens > 0


def test_openai_responses_api_builtin_tools_param():
    """Test that builtin_tools parameter is properly configured."""
    llm = OpenAICompletion(
        model="gpt-4o",
        api="responses",
        builtin_tools=["web_search", "code_interpreter"],
    )

    assert llm.builtin_tools == ["web_search", "code_interpreter"]

    messages = [{"role": "user", "content": "Test"}]
    params = llm._prepare_responses_params(messages)

    assert "tools" in params
    tool_types = [t["type"] for t in params["tools"]]
    assert "web_search_preview" in tool_types
    assert "code_interpreter" in tool_types


def test_openai_responses_api_builtin_tools_with_custom_tools():
    """Test that builtin_tools can be combined with custom function tools."""
    llm = OpenAICompletion(
        model="gpt-4o",
        api="responses",
        builtin_tools=["web_search"],
    )

    custom_tools = [
        {
            "name": "get_weather",
            "description": "Get weather for a location",
            "parameters": {"type": "object", "properties": {}},
        }
    ]

    messages = [{"role": "user", "content": "Test"}]
    params = llm._prepare_responses_params(messages, tools=custom_tools)

    assert len(params["tools"]) == 2
    tool_types = [t.get("type") for t in params["tools"]]
    assert "web_search_preview" in tool_types
    assert "function" in tool_types


@pytest.mark.vcr()
def test_openai_responses_api_with_web_search():
    """Test Responses API with web_search built-in tool."""
    llm = OpenAICompletion(
        model="gpt-4o-mini",
        api="responses",
        builtin_tools=["web_search"],
    )

    result = llm.call("What is the current population of Tokyo? Be brief.")

    assert isinstance(result, str)
    assert len(result) > 0


def test_responses_api_result_dataclass():
    """Test ResponsesAPIResult dataclass functionality."""
    result = ResponsesAPIResult(
        text="Hello, world!",
        response_id="resp_123",
    )

    assert result.text == "Hello, world!"
    assert result.response_id == "resp_123"
    assert result.web_search_results == []
    assert result.file_search_results == []
    assert result.code_interpreter_results == []
    assert result.computer_use_results == []
    assert result.reasoning_summaries == []
    assert result.function_calls == []
    assert not result.has_tool_outputs()
    assert not result.has_reasoning()


def test_responses_api_result_has_tool_outputs():
    """Test ResponsesAPIResult.has_tool_outputs() method."""
    result_with_web = ResponsesAPIResult(
        text="Test",
        web_search_results=[{"id": "ws_1", "status": "completed", "type": "web_search_call"}],
    )
    assert result_with_web.has_tool_outputs()

    result_with_file = ResponsesAPIResult(
        text="Test",
        file_search_results=[{"id": "fs_1", "status": "completed", "type": "file_search_call", "queries": [], "results": []}],
    )
    assert result_with_file.has_tool_outputs()


def test_responses_api_result_has_reasoning():
    """Test ResponsesAPIResult.has_reasoning() method."""
    result_with_reasoning = ResponsesAPIResult(
        text="Test",
        reasoning_summaries=[{"id": "r_1", "type": "reasoning", "summary": []}],
    )
    assert result_with_reasoning.has_reasoning()

    result_without = ResponsesAPIResult(text="Test")
    assert not result_without.has_reasoning()


def test_openai_responses_api_parse_tool_outputs_param():
    """Test that parse_tool_outputs parameter is properly configured."""
    llm = OpenAICompletion(
        model="gpt-4o",
        api="responses",
        parse_tool_outputs=True,
    )

    assert llm.parse_tool_outputs is True


def test_openai_responses_api_parse_tool_outputs_default_false():
    """Test that parse_tool_outputs defaults to False."""
    llm = OpenAICompletion(
        model="gpt-4o",
        api="responses",
    )

    assert llm.parse_tool_outputs is False


@pytest.mark.vcr()
def test_openai_responses_api_with_parse_tool_outputs():
    """Test Responses API with parse_tool_outputs enabled returns ResponsesAPIResult."""
    llm = OpenAICompletion(
        model="gpt-4o-mini",
        api="responses",
        builtin_tools=["web_search"],
        parse_tool_outputs=True,
    )

    result = llm.call("What is the current population of Tokyo? Be very brief.")

    assert isinstance(result, ResponsesAPIResult)
    assert len(result.text) > 0
    assert result.response_id is not None
    # Web search should have been used
    assert len(result.web_search_results) > 0
    assert result.has_tool_outputs()


@pytest.mark.vcr()
def test_openai_responses_api_parse_tool_outputs_basic_call():
    """Test Responses API with parse_tool_outputs but no built-in tools."""
    llm = OpenAICompletion(
        model="gpt-4o-mini",
        api="responses",
        parse_tool_outputs=True,
    )

    result = llm.call("Say hello in exactly 3 words.")

    assert isinstance(result, ResponsesAPIResult)
    assert len(result.text) > 0
    assert result.response_id is not None
    # No built-in tools used
    assert not result.has_tool_outputs()


# Auto-Chaining Tests (Responses API)


def test_openai_responses_api_auto_chain_param():
    """Test that auto_chain parameter is properly configured."""
    llm = OpenAICompletion(
        model="gpt-4o",
        api="responses",
        auto_chain=True,
    )

    assert llm.auto_chain is True
    assert llm._last_response_id is None


def test_openai_responses_api_auto_chain_default_false():
    """Test that auto_chain defaults to False."""
    llm = OpenAICompletion(
        model="gpt-4o",
        api="responses",
    )

    assert llm.auto_chain is False


def test_openai_responses_api_last_response_id_property():
    """Test last_response_id property."""
    llm = OpenAICompletion(
        model="gpt-4o",
        api="responses",
        auto_chain=True,
    )

    # Initially None
    assert llm.last_response_id is None

    # Simulate setting the internal value
    llm._last_response_id = "resp_test_123"
    assert llm.last_response_id == "resp_test_123"


def test_openai_responses_api_reset_chain():
    """Test reset_chain() method clears the response ID."""
    llm = OpenAICompletion(
        model="gpt-4o",
        api="responses",
        auto_chain=True,
    )

    llm._last_response_id = "resp_test_123"
    assert llm.last_response_id == "resp_test_123"

    llm.reset_chain()
    assert llm.last_response_id is None


def test_openai_responses_api_auto_chain_prepare_params():
    """Test that _prepare_responses_params uses auto-chained response ID."""
    llm = OpenAICompletion(
        model="gpt-4o",
        api="responses",
        auto_chain=True,
    )

    # No previous response ID yet
    params = llm._prepare_responses_params(messages=[{"role": "user", "content": "test"}])
    assert "previous_response_id" not in params

    llm._last_response_id = "resp_previous_123"
    params = llm._prepare_responses_params(messages=[{"role": "user", "content": "test"}])
    assert params.get("previous_response_id") == "resp_previous_123"


def test_openai_responses_api_explicit_previous_response_id_takes_precedence():
    """Test that explicit previous_response_id overrides auto-chained ID."""
    llm = OpenAICompletion(
        model="gpt-4o",
        api="responses",
        auto_chain=True,
        previous_response_id="resp_explicit_456",
    )

    llm._last_response_id = "resp_auto_123"

    # Explicit should take precedence
    params = llm._prepare_responses_params(messages=[{"role": "user", "content": "test"}])
    assert params.get("previous_response_id") == "resp_explicit_456"


def test_openai_responses_api_auto_chain_disabled_no_tracking():
    """Test that response ID is not tracked when auto_chain is False."""
    llm = OpenAICompletion(
        model="gpt-4o",
        api="responses",
        auto_chain=False,
    )

    # Even with a "previous" response ID set internally, params shouldn't use it
    llm._last_response_id = "resp_should_not_use"
    params = llm._prepare_responses_params(messages=[{"role": "user", "content": "test"}])
    assert "previous_response_id" not in params


@pytest.mark.vcr()
def test_openai_responses_api_auto_chain_integration():
    """Test auto-chaining tracks response IDs across calls."""
    llm = OpenAICompletion(
        model="gpt-4o-mini",
        api="responses",
        auto_chain=True,
    )

    assert llm.last_response_id is None
    result1 = llm.call("My name is Alice. Remember this.")

    # After first call, should have a response ID
    assert llm.last_response_id is not None
    first_response_id = llm.last_response_id
    assert first_response_id.startswith("resp_")

    result2 = llm.call("What is my name?")

    # Response ID should be updated
    assert llm.last_response_id is not None
    assert llm.last_response_id != first_response_id

    assert isinstance(result1, str)
    assert isinstance(result2, str)


@pytest.mark.vcr()
def test_openai_responses_api_auto_chain_with_reset():
    """Test that reset_chain() properly starts a new conversation."""
    llm = OpenAICompletion(
        model="gpt-4o-mini",
        api="responses",
        auto_chain=True,
    )

    llm.call("My favorite color is blue.")
    first_chain_id = llm.last_response_id
    assert first_chain_id is not None

    llm.reset_chain()
    assert llm.last_response_id is None

    llm.call("Hello!")
    second_chain_id = llm.last_response_id
    assert second_chain_id is not None
    assert second_chain_id != first_chain_id


# Encrypted Reasoning for ZDR (Zero Data Retention) Tests


def test_openai_responses_api_auto_chain_reasoning_param():
    """Test that auto_chain_reasoning parameter is properly configured."""
    llm = OpenAICompletion(
        model="gpt-4o",
        api="responses",
        auto_chain_reasoning=True,
    )

    assert llm.auto_chain_reasoning is True
    assert llm._last_reasoning_items is None


def test_openai_responses_api_auto_chain_reasoning_default_false():
    """Test that auto_chain_reasoning defaults to False."""
    llm = OpenAICompletion(
        model="gpt-4o",
        api="responses",
    )

    assert llm.auto_chain_reasoning is False


def test_openai_responses_api_last_reasoning_items_property():
    """Test last_reasoning_items property."""
    llm = OpenAICompletion(
        model="gpt-4o",
        api="responses",
        auto_chain_reasoning=True,
    )

    # Initially None
    assert llm.last_reasoning_items is None

    # Simulate setting the internal value
    mock_items = [{"id": "rs_test_123", "type": "reasoning"}]
    llm._last_reasoning_items = mock_items
    assert llm.last_reasoning_items == mock_items


def test_openai_responses_api_reset_reasoning_chain():
    """Test reset_reasoning_chain() method clears reasoning items."""
    llm = OpenAICompletion(
        model="gpt-4o",
        api="responses",
        auto_chain_reasoning=True,
    )

    # Set reasoning items
    mock_items = [{"id": "rs_test_123", "type": "reasoning"}]
    llm._last_reasoning_items = mock_items
    assert llm.last_reasoning_items == mock_items

    llm.reset_reasoning_chain()
    assert llm.last_reasoning_items is None


def test_openai_responses_api_auto_chain_reasoning_adds_include():
    """Test that auto_chain_reasoning adds reasoning.encrypted_content to include."""
    llm = OpenAICompletion(
        model="gpt-4o",
        api="responses",
        auto_chain_reasoning=True,
    )

    params = llm._prepare_responses_params(messages=[{"role": "user", "content": "test"}])
    assert "include" in params
    assert "reasoning.encrypted_content" in params["include"]


def test_openai_responses_api_auto_chain_reasoning_preserves_existing_include():
    """Test that auto_chain_reasoning preserves existing include items."""
    llm = OpenAICompletion(
        model="gpt-4o",
        api="responses",
        auto_chain_reasoning=True,
        include=["file_search_call.results"],
    )

    params = llm._prepare_responses_params(messages=[{"role": "user", "content": "test"}])
    assert "include" in params
    assert "reasoning.encrypted_content" in params["include"]
    assert "file_search_call.results" in params["include"]


def test_openai_responses_api_auto_chain_reasoning_no_duplicate_include():
    """Test that reasoning.encrypted_content is not duplicated if already in include."""
    llm = OpenAICompletion(
        model="gpt-4o",
        api="responses",
        auto_chain_reasoning=True,
        include=["reasoning.encrypted_content"],
    )

    params = llm._prepare_responses_params(messages=[{"role": "user", "content": "test"}])
    assert "include" in params
    assert params["include"].count("reasoning.encrypted_content") == 1


def test_openai_responses_api_auto_chain_reasoning_prepends_to_input():
    """Test that stored reasoning items are prepended to input."""
    llm = OpenAICompletion(
        model="gpt-4o",
        api="responses",
        auto_chain_reasoning=True,
    )

    # Simulate stored reasoning items
    mock_reasoning = MagicMock()
    mock_reasoning.type = "reasoning"
    mock_reasoning.id = "rs_test_123"
    llm._last_reasoning_items = [mock_reasoning]

    params = llm._prepare_responses_params(messages=[{"role": "user", "content": "test"}])

    assert len(params["input"]) == 2
    assert params["input"][0] == mock_reasoning
    assert params["input"][1]["role"] == "user"


def test_openai_responses_api_auto_chain_reasoning_disabled_no_include():
    """Test that reasoning.encrypted_content is not added when auto_chain_reasoning is False."""
    llm = OpenAICompletion(
        model="gpt-4o",
        api="responses",
        auto_chain_reasoning=False,
    )

    params = llm._prepare_responses_params(messages=[{"role": "user", "content": "test"}])
    assert "include" not in params or "reasoning.encrypted_content" not in params.get("include", [])


def test_openai_responses_api_auto_chain_reasoning_disabled_no_prepend():
    """Test that reasoning items are not prepended when auto_chain_reasoning is False."""
    llm = OpenAICompletion(
        model="gpt-4o",
        api="responses",
        auto_chain_reasoning=False,
    )

    # Even with stored reasoning items, they should not be prepended
    mock_reasoning = MagicMock()
    mock_reasoning.type = "reasoning"
    llm._last_reasoning_items = [mock_reasoning]

    params = llm._prepare_responses_params(messages=[{"role": "user", "content": "test"}])

    assert len(params["input"]) == 1
    assert params["input"][0]["role"] == "user"


def test_openai_responses_api_both_auto_chains_work_together():
    """Test that auto_chain and auto_chain_reasoning can be used together."""
    llm = OpenAICompletion(
        model="gpt-4o",
        api="responses",
        auto_chain=True,
        auto_chain_reasoning=True,
    )

    assert llm.auto_chain is True
    assert llm.auto_chain_reasoning is True
    assert llm._last_response_id is None
    assert llm._last_reasoning_items is None

    # Set both internal values
    llm._last_response_id = "resp_123"
    mock_reasoning = MagicMock()
    mock_reasoning.type = "reasoning"
    llm._last_reasoning_items = [mock_reasoning]

    params = llm._prepare_responses_params(messages=[{"role": "user", "content": "test"}])

    assert params.get("previous_response_id") == "resp_123"
    assert "reasoning.encrypted_content" in params["include"]
    assert len(params["input"]) == 2  # Reasoning item + message


# Agent Kickoff Structured Output Tests


@pytest.mark.vcr()
def test_openai_agent_kickoff_structured_output_without_tools():
    """
    Test that agent kickoff returns structured output without tools.
    This tests native structured output handling for OpenAI models.
    """
    from pydantic import BaseModel, Field

    class AnalysisResult(BaseModel):
        """Structured output for analysis results."""

        topic: str = Field(description="The topic analyzed")
        key_points: list[str] = Field(description="Key insights from the analysis")
        summary: str = Field(description="Brief summary of findings")

    agent = Agent(
        role="Analyst",
        goal="Provide structured analysis on topics",
        backstory="You are an expert analyst who provides clear, structured insights.",
        llm=LLM(model="gpt-4o-mini"),
        tools=[],
        verbose=True,
    )

    result = agent.kickoff(
        messages="Analyze the benefits of remote work briefly. Keep it concise.",
        response_format=AnalysisResult,
    )

    assert result.pydantic is not None, "Expected pydantic output but got None"
    assert isinstance(result.pydantic, AnalysisResult), f"Expected AnalysisResult but got {type(result.pydantic)}"
    assert result.pydantic.topic, "Topic should not be empty"
    assert len(result.pydantic.key_points) > 0, "Should have at least one key point"
    assert result.pydantic.summary, "Summary should not be empty"


@pytest.mark.vcr()
def test_openai_agent_kickoff_structured_output_with_tools():
    """
    Test that agent kickoff returns structured output after using tools.
    This tests post-tool-call structured output handling for OpenAI models.
    """
    from pydantic import BaseModel, Field
    from crewai.tools import tool

    class CalculationResult(BaseModel):
        """Structured output for calculation results."""

        operation: str = Field(description="The mathematical operation performed")
        result: int = Field(description="The result of the calculation")
        explanation: str = Field(description="Brief explanation of the calculation")

    @tool
    def add_numbers(a: int, b: int) -> int:
        """Add two numbers together and return the sum."""
        return a + b

    agent = Agent(
        role="Calculator",
        goal="Perform calculations using available tools",
        backstory="You are a calculator assistant that uses tools to compute results.",
        llm=LLM(model="gpt-4o-mini"),
        tools=[add_numbers],
        verbose=True,
    )

    result = agent.kickoff(
        messages="Calculate 15 + 27 using your add_numbers tool. Report the result.",
        response_format=CalculationResult,
    )

    assert result.pydantic is not None, "Expected pydantic output but got None"
    assert isinstance(result.pydantic, CalculationResult), f"Expected CalculationResult but got {type(result.pydantic)}"
    assert result.pydantic.result == 42, f"Expected result 42 but got {result.pydantic.result}"
    assert result.pydantic.operation, "Operation should not be empty"
    assert result.pydantic.explanation, "Explanation should not be empty"




def test_openai_stop_words_not_applied_to_structured_output():
    """
    Test that stop words are NOT applied when response_model is provided.
    This ensures JSON responses containing stop word patterns (like "Observation:")
    are not truncated, which would cause JSON validation to fail.
    """
    from pydantic import BaseModel, Field

    class ResearchResult(BaseModel):
        """Research result that may contain stop word patterns in string fields."""

        finding: str = Field(description="The research finding")
        observation: str = Field(description="Observation about the finding")

    # Create OpenAI completion instance with stop words configured
    llm = OpenAICompletion(
        model="gpt-4o",
        stop=["Observation:", "Final Answer:"],
    )

    # JSON response that contains a stop word pattern in a string field
    json_response = '{"finding": "The data shows growth", "observation": "Observation: This confirms the hypothesis"}'

    # This simulates what happens when the API returns JSON with stop word patterns
    result = llm._validate_structured_output(json_response, ResearchResult)

    assert isinstance(result, ResearchResult)
    assert result.finding == "The data shows growth"
    assert "Observation:" in result.observation


def test_openai_gpt5_models_do_not_support_stop_words():
    """
    Test that GPT-5 family models do not support stop words via the API.
    GPT-5 models reject the 'stop' parameter, so stop words must be
    applied client-side only.
    """
    gpt5_models = [
        "gpt-5",
        "gpt-5-mini",
        "gpt-5-nano",
        "gpt-5-pro",
        "gpt-5.1",
        "gpt-5.1-chat",
        "gpt-5.2",
        "gpt-5.2-chat",
    ]

    for model_name in gpt5_models:
        llm = OpenAICompletion(model=model_name)
        assert llm.supports_stop_words() == False, (
            f"Expected {model_name} to NOT support stop words"
        )


def test_openai_non_gpt5_models_support_stop_words():
    """
    Test that non-GPT-5 models still support stop words normally.
    """
    supported_models = [
        "gpt-4o",
        "gpt-4o-mini",
        "gpt-4.1",
        "gpt-4.1-mini",
        "gpt-4-turbo",
    ]

    for model_name in supported_models:
        llm = OpenAICompletion(model=model_name)
        assert llm.supports_stop_words() == True, (
            f"Expected {model_name} to support stop words"
        )


def test_openai_gpt5_still_applies_stop_words_client_side():
    """
    Test that GPT-5 models still truncate responses at stop words client-side
    via _apply_stop_words(), even though they don't send 'stop' to the API.
    """
    llm = OpenAICompletion(
        model="gpt-5.2",
        stop=["Observation:", "Final Answer:"],
    )

    assert llm.supports_stop_words() == False

    response = "I need to search.\n\nAction: search\nObservation: Found results"
    result = llm._apply_stop_words(response)

    assert "Observation:" not in result
    assert "Found results" not in result
    assert "I need to search" in result


@pytest.mark.parametrize(
    "model",
    ["gpt-5.6", "gpt-5.6-sol", "gpt-5.6-terra", "gpt-5.6-luna"],
)
def test_openai_gpt56_family_uses_official_context_window(model: str) -> None:
    """Native OpenAI must not inherit the shorter gpt-5 window for GPT-5.6."""
    llm = OpenAICompletion(model=model)
    assert llm.get_context_window_size() == int(1_050_000 * CONTEXT_WINDOW_USAGE_RATIO)


def test_openai_prefixed_gpt56_luna_uses_official_context_window() -> None:
    llm = LLM(model="openai/gpt-5.6-luna")
    assert isinstance(llm, OpenAICompletion)
    assert llm.model == "gpt-5.6-luna"
    assert llm.get_context_window_size() == int(1_050_000 * CONTEXT_WINDOW_USAGE_RATIO)


def test_openai_gpt5_and_gpt54_mini_keep_their_windows() -> None:
    gpt5 = OpenAICompletion(model="gpt-5")
    gpt54_mini = OpenAICompletion(model="gpt-5.4-mini")
    assert gpt5.get_context_window_size() == int(1_047_576 * CONTEXT_WINDOW_USAGE_RATIO)
    assert gpt54_mini.get_context_window_size() == int(200000 * CONTEXT_WINDOW_USAGE_RATIO)


def test_openai_stop_words_still_applied_to_regular_responses():
    """
    Test that stop words ARE still applied for regular (non-structured) responses.
    This ensures the fix didn't break normal stop word behavior.
    """
    # Create OpenAI completion instance with stop words configured
    llm = OpenAICompletion(
        model="gpt-4o",
        stop=["Observation:", "Final Answer:"],
    )

    # Response that contains a stop word - should be truncated
    response_with_stop_word = "I need to search for more information.\n\nAction: search\nObservation: Found results"

    result = llm._apply_stop_words(response_with_stop_word)

    # Response should be truncated at the stop word
    assert "Observation:" not in result
    assert "Found results" not in result
    assert "I need to search for more information" in result


def test_openai_structured_output_preserves_json_with_stop_word_patterns():
    """
    Test that structured output validation preserves JSON content
    even when string fields contain stop word patterns.
    """
    from pydantic import BaseModel, Field

    class AgentObservation(BaseModel):
        """Model with fields that might contain stop word-like text."""

        action_taken: str = Field(description="What action was taken")
        observation_result: str = Field(description="The observation result")
        final_answer: str = Field(description="The final answer")

    llm = OpenAICompletion(
        model="gpt-4o",
        stop=["Observation:", "Final Answer:", "Action:"],
    )

    # JSON that contains all the stop word patterns as part of the content
    json_with_stop_patterns = '''{
        "action_taken": "Action: Searched the database",
        "observation_result": "Observation: Found 5 relevant results",
        "final_answer": "Final Answer: The data shows positive growth"
    }'''

    # This should NOT be truncated since it's structured output
    result = llm._validate_structured_output(json_with_stop_patterns, AgentObservation)

    assert isinstance(result, AgentObservation)
    assert "Action:" in result.action_taken
    assert "Observation:" in result.observation_result
    assert "Final Answer:" in result.final_answer



@pytest.mark.vcr()
def test_openai_completions_cached_prompt_tokens():
    """
    Test that the Chat Completions API correctly extracts and tracks
    cached_prompt_tokens from prompt_tokens_details.cached_tokens.
    Sends the same large prompt twice so the second call hits the cache.
    """
    padding = "This is padding text to ensure the prompt is large enough for caching. " * 80
    system_msg = f"You are a helpful assistant. {padding}"

    llm = OpenAICompletion(model="gpt-4.1")

    llm.call([
        {"role": "system", "content": system_msg},
        {"role": "user", "content": "Say hello in one word."},
    ])

    llm.call([
        {"role": "system", "content": system_msg},
        {"role": "user", "content": "Say goodbye in one word."},
    ])

    usage = llm.get_token_usage_summary()
    assert usage.total_tokens > 0
    assert usage.prompt_tokens > 0
    assert usage.completion_tokens > 0
    assert usage.successful_requests == 2
    assert usage.cached_prompt_tokens > 0


@pytest.mark.vcr()
def test_openai_responses_api_cached_prompt_tokens():
    """
    Test that the Responses API correctly extracts and tracks
    cached_prompt_tokens from input_tokens_details.cached_tokens.
    """
    padding = "This is padding text to ensure the prompt is large enough for caching. " * 80
    system_msg = f"You are a helpful assistant. {padding}"

    llm = OpenAICompletion(model="gpt-4.1", api="responses")

    llm.call([
        {"role": "system", "content": system_msg},
        {"role": "user", "content": "Say hello in one word."},
    ])

    llm.call([
        {"role": "system", "content": system_msg},
        {"role": "user", "content": "Say goodbye in one word."},
    ])

    usage = llm.get_token_usage_summary()
    assert usage.total_tokens > 0
    assert usage.prompt_tokens > 0
    assert usage.completion_tokens > 0
    assert usage.successful_requests == 2
    assert usage.cached_prompt_tokens > 0


@pytest.mark.vcr()
def test_openai_streaming_cached_prompt_tokens():
    """
    Test that streaming Chat Completions API correctly extracts and tracks
    cached_prompt_tokens.
    """
    padding = "This is padding text to ensure the prompt is large enough for caching. " * 80
    system_msg = f"You are a helpful assistant. {padding}"

    llm = OpenAICompletion(model="gpt-4.1", stream=True)

    llm.call([
        {"role": "system", "content": system_msg},
        {"role": "user", "content": "Say hello in one word."},
    ])

    llm.call([
        {"role": "system", "content": system_msg},
        {"role": "user", "content": "Say goodbye in one word."},
    ])

    usage = llm.get_token_usage_summary()
    assert usage.total_tokens > 0
    assert usage.successful_requests == 2
    assert usage.cached_prompt_tokens > 0


@pytest.mark.vcr()
def test_openai_completions_cached_prompt_tokens_with_tools():
    """
    Test that the Chat Completions API correctly tracks cached_prompt_tokens
    when tools are used. The large system prompt should be cached across calls.
    """
    padding = "This is padding text to ensure the prompt is large enough for caching. " * 80
    system_msg = f"You are a helpful assistant that uses tools. {padding}"

    def get_weather(location: str) -> str:
        return f"The weather in {location} is sunny and 72°F"

    tools = [
        {
            "name": "get_weather",
            "description": "Get the current weather for a location",
            "parameters": {
                "type": "object",
                "properties": {
                    "location": {
                        "type": "string",
                        "description": "The city name"
                    }
                },
                "required": ["location"],
                "additionalProperties": False,
            },
        }
    ]

    llm = OpenAICompletion(model="gpt-4.1")

    llm.call(
        [
            {"role": "system", "content": system_msg},
            {"role": "user", "content": "What is the weather in Tokyo?"},
        ],
        tools=tools,
        available_functions={"get_weather": get_weather},
    )

    llm.call(
        [
            {"role": "system", "content": system_msg},
            {"role": "user", "content": "What is the weather in Paris?"},
        ],
        tools=tools,
        available_functions={"get_weather": get_weather},
    )

    usage = llm.get_token_usage_summary()
    assert usage.total_tokens > 0
    assert usage.prompt_tokens > 0
    assert usage.successful_requests == 2
    assert usage.cached_prompt_tokens > 0


@pytest.mark.vcr()
def test_openai_responses_api_cached_prompt_tokens_with_tools():
    """
    Test that the Responses API correctly tracks cached_prompt_tokens
    when function tools are used.
    """
    padding = "This is padding text to ensure the prompt is large enough for caching. " * 80
    system_msg = f"You are a helpful assistant that uses tools. {padding}"

    def get_weather(location: str) -> str:
        return f"The weather in {location} is sunny and 72°F"

    tools = [
        {
            "name": "get_weather",
            "description": "Get the current weather for a location",
            "parameters": {
                "type": "object",
                "properties": {
                    "location": {
                        "type": "string",
                        "description": "The city name"
                    }
                },
                "required": ["location"],
            },
        }
    ]

    llm = OpenAICompletion(model="gpt-4.1", api='responses')

    llm.call(
        [
            {"role": "system", "content": system_msg},
            {"role": "user", "content": "What is the weather in Tokyo?"},
        ],
        tools=tools,
        available_functions={"get_weather": get_weather},
    )

    llm.call(
        [
            {"role": "system", "content": system_msg},
            {"role": "user", "content": "What is the weather in Paris?"},
        ],
        tools=tools,
        available_functions={"get_weather": get_weather},
    )

    usage = llm.get_token_usage_summary()
    assert usage.total_tokens > 0
    assert usage.successful_requests == 2
    assert usage.cached_prompt_tokens > 0
def test_openai_streaming_returns_tool_calls_without_available_functions():
    """Test that streaming returns tool calls list when available_functions is None.

    This mirrors the non-streaming path where tool_calls are returned for
    the executor to handle. Reproduces the bug where streaming with tool
    calls would return empty text instead of tool_calls when
    available_functions was not provided (as the crew executor does).
    """
    llm = LLM(model="openai/gpt-4o-mini", stream=True)

    mock_chunk_1 = MagicMock()
    mock_chunk_1.choices = [MagicMock()]
    mock_chunk_1.choices[0].delta = MagicMock()
    mock_chunk_1.choices[0].delta.content = None
    mock_chunk_1.choices[0].delta.tool_calls = [MagicMock()]
    mock_chunk_1.choices[0].delta.tool_calls[0].index = 0
    mock_chunk_1.choices[0].delta.tool_calls[0].id = "call_abc123"
    mock_chunk_1.choices[0].delta.tool_calls[0].function = MagicMock()
    mock_chunk_1.choices[0].delta.tool_calls[0].function.name = "calculator"
    mock_chunk_1.choices[0].delta.tool_calls[0].function.arguments = '{"expr'
    mock_chunk_1.choices[0].finish_reason = None
    mock_chunk_1.usage = None
    mock_chunk_1.id = "chatcmpl-1"

    mock_chunk_2 = MagicMock()
    mock_chunk_2.choices = [MagicMock()]
    mock_chunk_2.choices[0].delta = MagicMock()
    mock_chunk_2.choices[0].delta.content = None
    mock_chunk_2.choices[0].delta.tool_calls = [MagicMock()]
    mock_chunk_2.choices[0].delta.tool_calls[0].index = 0
    mock_chunk_2.choices[0].delta.tool_calls[0].id = None
    mock_chunk_2.choices[0].delta.tool_calls[0].function = MagicMock()
    mock_chunk_2.choices[0].delta.tool_calls[0].function.name = None
    mock_chunk_2.choices[0].delta.tool_calls[0].function.arguments = 'ession": "1+1"}'
    mock_chunk_2.choices[0].finish_reason = None
    mock_chunk_2.usage = None
    mock_chunk_2.id = "chatcmpl-1"

    mock_chunk_3 = MagicMock()
    mock_chunk_3.choices = [MagicMock()]
    mock_chunk_3.choices[0].delta = MagicMock()
    mock_chunk_3.choices[0].delta.content = None
    mock_chunk_3.choices[0].delta.tool_calls = None
    mock_chunk_3.choices[0].finish_reason = "tool_calls"
    mock_chunk_3.usage = MagicMock()
    mock_chunk_3.usage.prompt_tokens = 10
    mock_chunk_3.usage.completion_tokens = 5
    mock_chunk_3.id = "chatcmpl-1"

    with patch.object(
        llm._client.chat.completions, "create", return_value=iter([mock_chunk_1, mock_chunk_2, mock_chunk_3])
    ):
        result = llm.call(
            messages=[{"role": "user", "content": "Calculate 1+1"}],
            tools=[{
                "type": "function",
                "function": {
                    "name": "calculator",
                    "description": "Calculate expression",
                    "parameters": {"type": "object", "properties": {"expression": {"type": "string"}}},
                },
            }],
            available_functions=None,
        )

    assert isinstance(result, list), f"Expected list of tool calls, got {type(result)}: {result}"
    assert len(result) == 1
    assert result[0]["function"]["name"] == "calculator"
    assert result[0]["function"]["arguments"] == '{"expression": "1+1"}'
    assert result[0]["id"] == "call_abc123"
    assert result[0]["type"] == "function"


def test_openai_responses_api_reasoning_tokens_extraction():
    """Test that reasoning_tokens are extracted from Responses API responses."""
    llm = LLM(model="openai/gpt-4o")

    mock_response = MagicMock()
    mock_response.usage = MagicMock(
        input_tokens=100,
        output_tokens=200,
        total_tokens=300,
    )
    mock_response.usage.input_tokens_details = MagicMock(cached_tokens=25)
    mock_response.usage.output_tokens_details = MagicMock(reasoning_tokens=80)

    usage = llm._extract_responses_token_usage(mock_response)
    assert usage["prompt_tokens"] == 100
    assert usage["completion_tokens"] == 200
    assert usage["total_tokens"] == 300
    assert usage["cached_prompt_tokens"] == 25
    assert usage["reasoning_tokens"] == 80


def test_openai_responses_api_no_detail_fields_omitted():
    """Test that reasoning/cached fields are omitted when Responses API details are absent."""
    llm = LLM(model="openai/gpt-4o")

    mock_response = MagicMock()
    mock_response.usage = MagicMock(
        input_tokens=50,
        output_tokens=30,
        total_tokens=80,
    )
    mock_response.usage.input_tokens_details = None
    mock_response.usage.output_tokens_details = None

    usage = llm._extract_responses_token_usage(mock_response)
    assert usage["prompt_tokens"] == 50
    assert usage["completion_tokens"] == 30
    assert "cached_prompt_tokens" not in usage
    assert "reasoning_tokens" not in usage


@pytest.mark.asyncio
async def test_openai_async_streaming_returns_tool_calls_without_available_functions():
    """Test that async streaming returns tool calls list when available_functions is None.

    Same as the sync test but for the async path (_ahandle_streaming_completion).
    """
    llm = LLM(model="openai/gpt-4o-mini", stream=True)

    mock_chunk_1 = MagicMock()
    mock_chunk_1.choices = [MagicMock()]
    mock_chunk_1.choices[0].delta = MagicMock()
    mock_chunk_1.choices[0].delta.content = None
    mock_chunk_1.choices[0].delta.tool_calls = [MagicMock()]
    mock_chunk_1.choices[0].delta.tool_calls[0].index = 0
    mock_chunk_1.choices[0].delta.tool_calls[0].id = "call_abc123"
    mock_chunk_1.choices[0].delta.tool_calls[0].function = MagicMock()
    mock_chunk_1.choices[0].delta.tool_calls[0].function.name = "calculator"
    mock_chunk_1.choices[0].delta.tool_calls[0].function.arguments = '{"expr'
    mock_chunk_1.choices[0].finish_reason = None
    mock_chunk_1.usage = None
    mock_chunk_1.id = "chatcmpl-1"

    mock_chunk_2 = MagicMock()
    mock_chunk_2.choices = [MagicMock()]
    mock_chunk_2.choices[0].delta = MagicMock()
    mock_chunk_2.choices[0].delta.content = None
    mock_chunk_2.choices[0].delta.tool_calls = [MagicMock()]
    mock_chunk_2.choices[0].delta.tool_calls[0].index = 0
    mock_chunk_2.choices[0].delta.tool_calls[0].id = None
    mock_chunk_2.choices[0].delta.tool_calls[0].function = MagicMock()
    mock_chunk_2.choices[0].delta.tool_calls[0].function.name = None
    mock_chunk_2.choices[0].delta.tool_calls[0].function.arguments = 'ession": "1+1"}'
    mock_chunk_2.choices[0].finish_reason = None
    mock_chunk_2.usage = None
    mock_chunk_2.id = "chatcmpl-1"

    mock_chunk_3 = MagicMock()
    mock_chunk_3.choices = [MagicMock()]
    mock_chunk_3.choices[0].delta = MagicMock()
    mock_chunk_3.choices[0].delta.content = None
    mock_chunk_3.choices[0].delta.tool_calls = None
    mock_chunk_3.choices[0].finish_reason = "tool_calls"
    mock_chunk_3.usage = MagicMock()
    mock_chunk_3.usage.prompt_tokens = 10
    mock_chunk_3.usage.completion_tokens = 5
    mock_chunk_3.id = "chatcmpl-1"

    class MockAsyncStream:
        """Async iterator that mimics OpenAI's async streaming response."""

        def __init__(self, chunks: list[Any]) -> None:
            self._chunks = chunks
            self._index = 0

        def __aiter__(self) -> "MockAsyncStream":
            return self

        async def __anext__(self) -> Any:
            if self._index >= len(self._chunks):
                raise StopAsyncIteration
            chunk = self._chunks[self._index]
            self._index += 1
            return chunk

    async def mock_create(**kwargs: Any) -> MockAsyncStream:
        return MockAsyncStream([mock_chunk_1, mock_chunk_2, mock_chunk_3])

    with patch.object(
        llm._async_client.chat.completions, "create", side_effect=mock_create
    ):
        result = await llm.acall(
            messages=[{"role": "user", "content": "Calculate 1+1"}],
            tools=[{
                "type": "function",
                "function": {
                    "name": "calculator",
                    "description": "Calculate expression",
                    "parameters": {"type": "object", "properties": {"expression": {"type": "string"}}},
                },
            }],
            available_functions=None,
        )

    assert isinstance(result, list), f"Expected list of tool calls, got {type(result)}: {result}"
    assert len(result) == 1
    assert result[0]["function"]["name"] == "calculator"
    assert result[0]["function"]["arguments"] == '{"expression": "1+1"}'
    assert result[0]["id"] == "call_abc123"
    assert result[0]["type"] == "function"


def test_openai_reasoning_tokens_extraction():
    """Test that reasoning_tokens are extracted from OpenAI o-series responses."""
    llm = LLM(model="openai/gpt-4o")

    mock_response = MagicMock()
    mock_response.usage = MagicMock(
        prompt_tokens=100,
        completion_tokens=200,
        total_tokens=300,
    )
    mock_response.usage.prompt_tokens_details = MagicMock(cached_tokens=25)
    mock_response.usage.completion_tokens_details = MagicMock(reasoning_tokens=80)

    usage = llm._extract_openai_token_usage(mock_response)
    assert usage["prompt_tokens"] == 100
    assert usage["completion_tokens"] == 200
    assert usage["total_tokens"] == 300
    assert usage["cached_prompt_tokens"] == 25
    assert usage["reasoning_tokens"] == 80


def test_openai_no_detail_fields_omitted():
    """Test that reasoning/cached fields are omitted when details are absent."""
    llm = LLM(model="openai/gpt-4o")

    mock_response = MagicMock()
    mock_response.usage = MagicMock(
        prompt_tokens=50,
        completion_tokens=30,
        total_tokens=80,
    )
    mock_response.usage.prompt_tokens_details = None
    mock_response.usage.completion_tokens_details = None

    usage = llm._extract_openai_token_usage(mock_response)
    assert usage["prompt_tokens"] == 50
    assert usage["completion_tokens"] == 30
    assert "cached_prompt_tokens" not in usage
    assert "reasoning_tokens" not in usage